Randomized, Phase III Trial of First-Line Figitumumab in
Combination With Paclitaxel and Carboplatin Versus
Paclitaxel and Carboplatin Alone in Patients With Advanced
Non–Small-Cell Lung Cancer
Corey J. Langer, Silvia Novello, Keunchil Park, Maciej Krzakowski, Daniel D. Karp, Tony Mok, Rebecca J. Benner, Judith R. Scranton, Anthony J. Olszanski, and Jacek Jassem
Corey J. Langer, University of Pennsyl-vania; Anthony J. Olszanski, Fox Chase Cancer Center, Philadelphia, PA; Daniel D. Karp, MD Anderson Center, Hous-ton, TX; Rebecca J. Benner, Judith R. Scranton, Pfizer Oncology, Groton, CT; Silvia Novello, University of Turin, Orbassano, Italy; Keunchil Park, Sung-kyunkwan University School of Medi-cine, Seoul, Korea; Maciej Krzakowski, The Maria Sklodowska-Curie Institute of Oncology, Warsaw; Jacek Jassem, Medical University of Gdansk, Gdansk, Poland; Tony Mok, Chinese University, Hong Kong, Special Administrative Region, People’s Republic of China. Published online ahead of print at www.jco.org on June 2, 2014. Supported by Pfizer and by Grant No. CA16672 from the National Cancer Institute (Clinical and Translational Research Center).
Presented in part at the 46th Annual Meeting of the American Society of Clinical Oncology, Chicago, IL, June 4-8, 2010.
Authors’ disclosures of potential con-flicts of interest and author contribu-tions are found at the end of this article.
Clinical trial information: NCT00596830. Corresponding author: Corey J. Langer, MD, FACP, Professor of Medicine, Hematology-Oncology Division, Univer-sity of Pennsylvania, 3400 Civic Center Blvd, 2 Perelman Center for Advanced Medicine, Philadelphia, PA 19104; e-mail: corey.langer@uphs.upenn.edu. © 2014 by American Society of Clinical Oncology
0732-183X/14/3219w-2059w/$20.00 DOI: 10.1200/JCO.2013.54.4932
A B S T R A C T
Purpose
Figitumumab (CP-751,871), a fully human immunoglobulin G2 monoclonal antibody, inhibits the insulin-like growth factor 1 receptor (IGF-1R). Our multicenter, randomized, phase III study compared figitumumab plus chemotherapy with chemotherapy alone as first-line treatment in patients with advanced non–small-cell lung cancer (NSCLC).
Patients and Methods
Patients with stage IIIB/IV or recurrent NSCLC disease with nonadenocarcinoma histology received open-label figitumumab (20 mg/kg) plus paclitaxel (200 mg/m2) and carboplatin (area
under the concentration-time curve, 6 mg䡠 min/mL) or paclitaxel and carboplatin alone once every 3 weeks for up to six cycles. The primary end point was overall survival (OS).
Results
Of 681 randomly assigned patients, 671 received treatment. The study was closed early by an independent Data Safety Monitoring Committee because of futility and an increased incidence of serious adverse events (SAEs) and treatment-related deaths with figitumumab. Median OS was 8.6 months for figitumumab plus chemotherapy and 9.8 months for chemotherapy alone (hazard ratio [HR], 1.18; 95% CI, 0.99 to 1.40; P⫽ .06); median progression-free survival was 4.7 months (95% CI, 4.2 to 5.4) and 4.6 months (95% CI, 4.2 to 5.4), respectively (HR, 1.10; P⫽ .27); the objective response rates were 33% and 35%, respectively. The respective rates of all-causality SAEs were 66% and 51%; P⬍ .01). Treatment-related grade 5 adverse events were also more common with figitumumab (5% v 1%; P⬍ .01).
Conclusion
Adding figitumumab to standard chemotherapy failed to increase OS in patients with advanced nonadenocarcinoma NSCLC. Further clinical development of figitumumab is not being pursued.
J Clin Oncol 32:2059-2066. © 2014 by American Society of Clinical Oncology
INTRODUCTION
Metastatic non–small-cell lung cancer (NSCLC) is rarely curable and, despite significant treatment ad-vances over the last decade, 5-year survival rates remain below 5%.1 Current therapeutic options include histology-based chemotherapy, antiangio-genic agents, and targeted agents inhibiting epidermal growth factor receptor and anaplastic lymphoma kinase. Insulin-like growth factor 1 (IGF-1) receptor (IGF-1R) is a central component of cancer signal transduction pathways.2Expression of IGF-1R is detectable in 39% to 84% of advanced NSCLCs and is more frequently found in squamous
cell lung cancer.3 The prognostic significance of IGF-1R expression remains unclear. Several pro-spective studies suggest a relationship between cir-culating IGF-1 and cancer risk.4,5
Figitumumab (CP-751,871) is a fully human immunoglobulin G2 monoclonal antibody that in-hibits IGF-1R. In phase I trials, it was well tolerated as a single agent and in combination with chemo-therapy at 20 mg/kg every 3 weeks.6,7In a random-ized phase II study of patients with treatment-naive advanced NSCLC, the originally reported objective response rate (ORR) was 54% with figitumumab 10 or 20 mg/kg plus full-dose paclitaxel and carbopla-tin, and 42% with chemotherapy alone. Median
progression-free survival (PFS) was initially reported as 5.0 months with figitumumab 20 mg/kg and 3.5 months with chemotherapy alone. No unexpected toxicities were observed. These findings prompted a prospective, randomized phase III trial of figitumumab plus paclitaxel and carboplatin compared with chemotherapy alone as first-line treatment for advanced NSCLC. However, the phase II data were subsequently retracted after a reanalysis revealed a lower ORR in both treatment arms (see Discussion).8
In this article, we report the results of the phase III trial, which was restricted to patients with nonadenocarcinoma histology based on an initial analysis of the phase II study that indicated potentially increased figitumumab efficacy in this subset.8
PATIENTS AND METHODS
Patients
Eligible patients were at least 18 years old with histologically or cytolog-ically confirmed advanced NSCLC; documented American Joint Committee on Cancer9stage IIIB or metastatic (stage IV or recurrent) disease not amena-ble to curative treatment; and a primary histology of predominantly squamous cell, large cell, or adenosquamous carcinoma. Prior systemic treatment for NSCLC and previous or concurrent therapy with IGF-1R inhibitors or growth hormone agonists or antagonists were prohibited. Adjuvant chemotherapy was permitted if completed at least 12 months before randomization. Prior surgery or radiation therapy was permitted if completed at least 3 weeks before randomization, with all acute toxicities resolved to National Cancer Institute Common Terminology Criteria for Adverse Events version 3.0 grade 1. Pa-tients had to have an Eastern Cooperative Oncology Group performance status of 0 or 1 and adequate organ function. Exclusions included symptom-atic CNS metastases, other active malignancies, uncontrolled hypertension, or uncontrolled diabetes (baseline glycosylated hemoglobin [HbA1c]⬎ 8%).
The study was conducted in accordance with International Conference on Harmonisation Good Clinical Practice guidelines, the declaration of Hel-sinki, and local regulatory requirements and laws. Institutional review board or independent ethics committee approval was required for each investigator and center. Written informed consent was obtained from all patients. Study Design and Treatment
Patients were randomly assigned in a 1:1 ratio to open-label figitu-mumab plus paclitaxel and carboplatin (investigational arm) or paclitaxel and carboplatin alone (control arm), stratified by previous adjuvant chem-otherapy, sex, and histology (squamous-cell v combined large-cell or ade-nosquamous cancer).
The primary end point was overall survival (OS), which was defined as time from randomization to death as a result of any cause. Secondary end points included PFS, ORR, and safety. The association between serum IGF-1 levels and OS was a preplanned exploratory objective.
All patients received carboplatin (area under the concentration-time curve, 6 mg䡠min/mL)andpaclitaxel(200mg/m2) intravenously on day 1 once every 3 weeks for up to six cycles. In the investigational arm, patients also received figitumumab 20 mg/kg intravenously on day 1 of each 3-week cycle, for up to 17 cycles (approximately 1 year of treatment).
On the investigational arm, if paclitaxel and/or carboplatin were discon-tinued early, patients could continue single-agent figitumumab (once every 3 weeks) until disease progression or intolerance. Additional cycles were permit-ted in patients exhibiting response, based on agreement between the study sponsor and investigator. If figitumumab was discontinued, paclitaxel and carboplatin were continued for a maximum of six cycles until disease progres-sion or intolerance. Standard supportive therapies were instituted in both arms. Guidelines for managing emergent hyperglycemia were provided, including immediate treatment, protocol-defined figitumumab-dosage modification, and continued oral glucose-lowering therapy if hyperglycemia was expected to continue.
Study Procedures
Tumor assessment was performed at baseline and every 6 weeks until radiologic disease progression or initiation of subsequent anticancer therapy using Response Evaluation Criteria in Solid Tumors version 1.0.10Adverse events were graded using National Cancer Institute Common Terminology Criteria for Adverse Events version 3.0. Clinical assessments, including hema-tology and serum chemistry, were performed at baseline, on day 1 of cycle 1 (all measurements), days 8 and 15 of cycle 1 (hematology only), on day 1 of each subsequent cycle, and at the end of treatment. Levels of HbA1cwere measured at baseline, before cycle 4, and at the end of treatment.
Serum samples were collected within 2 hours before chemotherapy and/or figitumumab infusion at cycles 1 and 4 and at the end of treatment. Total IGF-1 levels were determined by immunochemiluminometric assay at MDS Pharma Services (now LabCorp; Mississauga, Ontario, Canada). An independent Data Safety Monitoring Committee (DSMC) monitored safety and efficacy.
Statistical Analysis
With one-sided .025 level testing and 90% power, 820 patients were needed to detect a 30% improvement for figitumumab plus chemotherapy over the median 10-month survival rate seen with paclitaxel plus carboplatin therapy (hazard ratio [HR], 0.77); 649 events were expected at full follow-up. The primary assessment was a log-rank test stratified by factors used in ran-domization. The analysis set included all randomly assigned patients on an intent-to-treat basis. Two-sided P values were determined.
Two interim analyses were planned after approximately one third and two thirds of the anticipated number of events had occurred. A Lan-DeMets spending function approach with O’Brien-Fleming stopping bounds (Appen-dix Table A1 [online-only]) was used to reject the null hypothesis (efficacy boundary) and the alternative hypothesis (futility boundary). Statistical anal-yses were conducted by Pfizer.
RESULTS
Patients and Treatment Exposure
Between April 2008 and September 2009, 681 patients from 163 sites in 25 countries were randomly assigned and 671 received treat-ment (figitumumab group, 338; control group, 333; Fig 1). Demo-graphic and baseline characteristics were well balanced between treatment arms (Table 1). Patients’ median age was 62 years. Most of the patients were men and most had stage IV disease. Patients in the figitumumab and control arms received a median of four and five cycles of chemotherapy, respectively (Table 2); 33% and 44%, respec-tively, completed six cycles of paclitaxel, and 34% and 46% completed six cycles of carboplatin. Figitumumab-treated patients received a median of four figitumumab cycles; 109 (32%) of 338 figitumumab-treated patients received four to six cycles, seven (2%) of 338 patients received 17 cycles, and four (1%) of 338 received more than 20 cycles. A total of 124 (37%) of 338 patients received figitumumab after completing or discontinuing chemotherapy (median of two mainte-nance cycles). Of these, 87 (26%) of 338 patients received figitu-mumab maintenance after six cycles (maintenance therapy could start earlier than cycle 6).
On DSMC advice, enrollment was suspended in September 2009 because of a higher number of serious adverse events (SAEs) and deaths in the figitumumab arm. The study was permanently closed to new accrual in December 2009, after the first interim analysis indi-cated that the addition of figitumumab was highly unlikely to meet the primary end point of improving OS over chemotherapy alone. Follow-up for OS continued until March 2011. The overall median follow-up time was 23.1 months (Table 3).
Efficacy
At the final analysis, 259 patients in the figitumumab arm and 251 in the control arm had died (Table 3). The median OS was 8.6 months (95% CI, 7.4 to 9.3) and 9.8 months (95% CI, 8.6 to 10.9), respectively (HR, 1.18; 95% CI, 0.99 to 1.40; P⫽ .06; Fig 2A). Respec-tive 1-year survival rates were 34% and 39%. The effect of figitu-mumab was similar across all subgroups based on demographic or other baseline characteristics (Fig 3).
Median PFS was 4.7 months for the figitumumab arm (95% CI, 4.2 to 5.4) and 4.6 months for the control arm (95% CI, 4.2 to 5.4; HR, 1.10; 95% CI, 0.93 to 1.32; P⫽ .27; Fig 2B). Respective ORRs were 33% (95% CI, 28 to 38) and 35% (95% CI, 29 to 40; Table 3).
Safety
Alopecia and nausea were the most common treatment-emergent (all-causality) adverse events (AEs) of any grade and occurred in a similar number of patients in each arm (Table 4).
Any-grade AEs that occurred more frequently in the figitumumab arm included hyperglycemia, diarrhea, decreased appetite, vomiting, and decreased weight. Grade 3/4 AEs that occurred more frequently in the figitumumab arm included hyperglycemia, decreased appetite, dehydration, diarrhea, fatigue, and nausea.
Treatment-emergent (all-causality) SAEs occurred in 66% of the figitumumab arm and 51% of the control arm (P⬍ .01 by Fisher’s exact test). Excluding disease progression, the most common SAEs were pneumonia (6% v 4%, respectively), dehydration (4% v 1%), asthenia (3% v 1%), and hyperglycemia (3% v⬍ 1%). The SAEs were judged to have a reasonable possibility of being treatment-related in 22% and 12% of patients, respectively.
Nonprogression grade 5 AEs occurred in 13% of the figitu-mumab arm and 10% of the control arm (P⫽ .22). The most com-mon grade 5 AEs in the figitumumab arm were pulcom-monary hemorrhage and pneumonia (2% each; Appendix Table A2). Grade 5 AEs were considered to be treatment-related in 5% of the Randomly assigned, stratified by previous adjuvant chemotherapy,
sex, and histology (squamous cell v combined large cell or adenosquamous) (N = 681)
Figitumumab arm
Allocated to figitumumab plus chemotherapy (n = 342) Received figitumumab plus chemotherapy (n = 338) Did not receive figitumumab plus chemotherapy (n = 3) * ) 1 = n ( y p a r e h t o m e h c y l n o d e v i e c e R Control arm ) 9 3 3 = n ( y p a r e h t o m e h c o t d e t a c o l l A ) 2 3 3 = n ( y p a r e h t o m e h c d e v i e c e R ) 7 = n ( y p a r e h t o m e h c e v i e c e r t o n d i D ) 2 4 3 = n ( y c a c i f f e r o f d e z y l a n A
Analyzed for safety
) 8 3 3 = n ( s t n e v e e s r e v d A ) 4 3 3 = n ( a t a d y r o t a r o b a L ) 4 = n ( f f o t u c a t a d r e t f a p u -w o l l o F ) 2 = n ( b a m u m u t i g i f d e u n i t n o c s i D ) 1 = n ( h t a e d d n a n o i s s e r g o r p e v i t c e j b O
Transition to off-study, single-patient IND (n = 1)
) 9 3 3 = n ( y c a c i f f e r o f d e z y l a n A
Analyzed for safety
) 3 3 3 = n ( s t n e v e e s r e v d A ) 4 2 3 = n ( a t a d y r o t a r o b a L Receiving figitumumab at time of data cutoff (n = 2)
Receiving chemotherapy at time of data cutoff (n = 0) ) 6 3 3 = n ( b a m u m u t i g i f d e u n i t n o c s i D ) 3 4 = n ( h t a e D ) 2 2 = n ( b a m u m u t i g i f o t d e t a l e r E A ) 2 3 = n ( b a m u m u t i g i f o t d e t a l e r n u E A ) 0 1 = n ( n o i t a r o i r e t e d h t l a e h l a b o l G ) 2 = n ( p u -w o l l o f o t t s o L ) 0 7 1 = n ( e s p a l e r r o n o i s s e r g o r p e v i t c e j b O ) 2 2 = n ( r e h t O ) 1 = n ( n o i t a l o i v l o c o t o r P ) 4 3 = n ( E A o t d e t a l e r n u l a s u f e r t n e i t a P ) 9 0 2 = n ( y p a r e h t o m e h c d e u n i t n o c s i D ) 0 4 = n ( h t a e D ) 9 2 = n ( y p a r e h t o m e h c o t d e t a l e r E A ) 7 2 = n ( y p a r e h t o m e h c o t d e t a l e r n u E A ) 9 = n ( n o i t a r o i r e t e d h t l a e h l a b o l G ) 2 = n ( p u -w o l l o f o t t s o L ) 8 7 = n ( e s p a l e r r o n o i s s e r g o r p e v i t c e j b O ) 5 = n ( r e h t O ) 1 = n ( n o i t a l o i v l o c o t o r P ) 8 1 = n ( E A o t d e t a l e r n u l a s u f e r t n e i t a P ) 9 2 1 = n ( y p a r e h t o m e h c d e t e l p m o C
Receiving chemotherapy at time of data cutoff (n = 0) ) 1 = n ( b a m u m u t i g i f d e u n i t n o c s i D
AE unrelated to figitumumab (patient was randomly (n = 1) assigned to figitumumab arm but received only
paclitaxel and carboplatin)
) 6 8 1 = n ( y p a r e h t o m e h c d e u n i t n o c s i D ) 2 3 = n ( h t a e D ) 1 3 = n ( y p a r e h t o m e h c o t d e t a l e r E A ) 3 1 = n ( y p a r e h t o m e h c o t d e t a l e r n u E A ) 0 1 = n ( n o i t a r o i r e t e d h t l a e h l a b o l G ) 1 = n ( p u -w o l l o f o t t s o L ) 1 8 = n ( e s p a l e r r o n o i s s e r g o r p e v i t c e j b O ) 8 = n ( r e h t O ) 1 = n ( n o i t a l o i v l o c o t o r P ) 9 = n ( E A o t d e t a l e r n u l a s u f e r t n e i t a P ) 7 4 1 = n ( y p a r e h t o m e h c d e t e l p m o C
figitumumab arm and 1% of the control arm (P⬍ .01). With figitu-mumab, these grade 5 AEs included hemoptysis, pneumonia, un-known cause reported only as death, septic shock, cardiorespiratory arrest, decrease of performance status, neutropenic sepsis, toxicity to various agents, renal failure, hemorrhage, and hypovolemic shock (ⱕ 1% each). In the control arm, the grade 5 AEs included unknown cause reported as death, pneumonia, septic shock, and dehydration (⬍ 1% each).
Figitumumab was discontinued because of treatment-related AEs in 7% of patients, and chemotherapy was discontinued for this reason in 9% of patients in each arm.
Relationship of Total IGF-1 and HbA1cto Outcomes
For the exploratory analysis of outcomes based on baseline total IGF-1, a cutoff of 120 ng/mL was selected because it was associated
with the largest observed differences in treatment effect above and below it. Baseline IGF-1 was not related to overall frequency or nature of AEs. However, grade 5 AEs were more common among figitumumab-treated patients with baseline IGF-1 levels less than 120 ng/mL (56%) than among those with baseline levels of 120 ng/mL or higher (38%) and those in the control arm (37% and 36% in the low and high IGF-1 groups, respectively). In the figitumumab arm, me-dian OS for patients with low and high baseline IGF-1 was 7.0 months and 10.4 months, respectively; in the control arm it was 10.1 and 9.4 months, respectively (Appendix Fig A1). For patients with high IGF-1, there was no difference in OS between treatment groups (HR, 0.93; P⫽ .67). For those patients with low IGF-1, OS was significantly shorter in the figitumumab arm (HR, 1.37; P⫽ .01).
The rate of all-causality AEs did not vary markedly by baseline HbA1cstatus, but the rate of grade 3/4 AEs for patients with no grade 5 events was slightly lower in those with baseline levels less than 5.7% than in those with levelsⱖ 5.7% (figitumumab arm, 30% v 36%; control arm, 33% v 35%). Median OS in patients with low baseline HbA1cwas 8.7 months in the figitumumab arm and 10.2 months in the control arm (HR, 1.07; P⫽ .65). The respective values in patients with high HbA1cwere 8.2 and 9.7 months (HR, 1.26; P⫽ .05).
DISCUSSION
This was the first randomized phase III study to test whether combin-ing an IGF-1R inhibitor (figitumumab) with paclitaxel and carbopla-tin could improve OS versus chemotherapy alone as first-line treatment for advanced nonadenocarcinoma NSCLC. When this trial was initiated, IGF-1R was thought to play an important role in squa-mous cell histology NSCLC, an area of particular unmet need. Unex-pectedly, adding figitumumab to chemotherapy proved deleterious. The DSMC closed the study because of therapeutic futility and in-creased SAEs, including treatment-related deaths, in the figitumumab arm. This outcome was disappointing given the originally reported phase II ORR of 54% for combination therapy compared with 42% for chemotherapy alone.8
The phase III study was designed and conducted in good faith based on the aforementioned phase II trial findings in treatment-naive advanced NSCLC. Following closure of the phase III trial, the phase II data were retracted after a reanalysis revealed a lower ORR in both treatment arms.8In addition, median PFS no longer trended in favor of figitumumab (4.5 months with figitumumab 20 mg/kg and 4.3 months with chemotherapy alone). The heightened toxicity of figitu-mumab in the phase III trial was not observed in the original phase I/II trials in NSCLC, which enrolled more than 150 patients in total. In our current study, the figitumumab combination failed to improve any efficacy end points over chemotherapy alone. Overall survival, the primary end point, was 8.6 months versus 9.8 months respectively. The ORR with figitumumab (33%) was similar to that observed in the phase II final analysis (37% in both the overall cohort [initially re-ported as 54%] and the nonadenocarcinoma cohort). Another ad-vanced NSCLC trial, initiated after the phase III was underway, used the same treatment in combination with figitumumab and the ORR was 39%.12
Subgroup analysis suggests that figitumumab safety and tolera-bility were poorer in patients with low baseline IGF-1 (⬍ 120 ng/mL) compared with those with high IGF-1 (ⱖ 120 ng/mL), particularly
Table 1. Baseline Patient Characteristics
Characteristic Figitumumab Arm (n⫽ 342) Control Arm(n⫽ 339) No. of Patients % No. of Patients % Sex Male 261 76 260 77 Female 81 24 79 23 Age, years Median 62 62 Range 30-90 36-83 Ethnicity White 265 78 270 80 Asian 56 16 59 17 Black 9 3 4 1 Other 12 4 6 2
ECOG performance status
0 113 33 115 34
1 226 66 217 64
Not reported 3 1 7 2
Current disease stageⴱ
Stage IIIB 39 11 39 12 Stage IV 302 88 300 88 Not reported 1 ⬍ 1 0 Smoking status Never smoked 34 10 33 10 Current smoker 142 42 141 42 Former smoker 166 49 165 49 Histology Squamous cell 295 86 289 85 Large cell 28 8 26 8 Adenosquamous 15 4 19 6 Other 4 1 5 1 Prior treatments Surgery 72 21 61 18 Radiation 44 13 36 11 Adjuvant chemotherapy† 14 4 15 4
NOTE. Data are presented for all patients by randomized arm. The stratification factors (histology, sex, and adjuvant chemotherapy) are presented as collected in the case report forms rather than as collected by the randomization system.
Abbreviation: ECOG, Eastern Cooperative Oncology Group. ⴱTNM Classification of Malignant Tumours (6th ed).11
†Systemic therapy included carboplatin/paclitaxel (n⫽ 3), cisplatin/vinorel-bine (n⫽ 7), cisplatin/gemcitabine (n ⫽ 4), carboplatin/gemcitabine (n ⫽ 3), cisplatin/etoposide (n⫽ 3), other regimens with carboplatin (n ⫽ 3), other regimens with cisplatin (n⫽ 4), and other nonplatinum regimens (n ⫽ 2).
with respect to grade 5 AEs. Consequently, in the figitumumab arm, median OS was shorter in patients with low IGF-1 compared with those with high IGF-1 and was significantly shorter compared with control patients who had low IGF-1 (HR, 1.37, P⫽ .01). Although additional studies are required, these data suggest that low baseline total IGF-1 may be a safety biomarker that identifies a subset of patients for whom IGF-1R inhibition is particularly harmful. In a phase I study of ganitumab, a human monoclonal antibody against IGF-1R, treatment transiently increased IGF-1.13Low baseline IGF-1 may indicate an inability to mount a compensatory increase in IGF-1 and greater likelihood of AEs.
Hyperglycemia of any grade occurred more frequently in the figitu-mumab arm than in the control arm (23% v 5%), as did grade 3/4 hyperglycemia(12%v1%).Hyperglycemiawasoneofthemostcommon SAEs, with greater frequency in the figitumumab arm than in the control arm. Hyperglycemia is likely a class effect stemming from impaired ho-meostatic control of glucose metabolism as a consequence of IGF-1R inhibition.14Hyperglycemia was rarely severe and was usually manage-able with agents such as metformin, but could have contributed in subtle ways to increased toxicity in the figitumumab arm.
Baseline HbA1c was not a strong biosafety marker, although grade 3/4 AEs were slightly more common in patients with levels
Table 2. Study Drug Exposure
Treatment Delivery
Figitumumab Arm (n⫽ 338) Control Arm (n⫽ 333)ⴱ Figitumumab Paclitaxel Carboplatin Paclitaxel Carboplatin No. of Patients % No. of Patients % No. of Patients % No. of Patients % No. of Patients % Cycles started Median 4 4 4 5 5 Range 1-52† 1-6 1-6 1-6 1-6
Duration of treatment, weeks
Median 12.1 10.2 10.2 12.3 12.6
Range 0.1-161.3† 0.1-20.1 0.1-20.1 0.1-23 0.1-23
Patients with at least one dose reduction 30 9 55 16 51 15 52 16 44 13
Patients with at least one dose delay 76 22 52 15 54 16 42 13 42 13
Abbreviation: IND, investigational new drug.
ⴱPatients in the chemotherapy arm who received carboplatin, n⫽ 332.
†Includes experience after data cutoff from the last ongoing patient, who transitioned to single patient IND in September 2012.
Table 3. Efficacy Results
End Point
Figitumumab Arm (n⫽ 342) Control Arm (n⫽ 339)
Hazard Ratio P No. of Patients % No. of Patients %
Median follow-up time, months 23.3 22.8
Overall survival (primary end point) .06ⴱ
Patient deaths 259 76 251 74
Median, months† 8.6 9.8 1.18‡
95% CI 7.4 to 9.3 8.6 to 10.9 0.99 to 1.40
One-year survival§ 34 39
Progression-free survival (investigator assessment) .27ⴱ
Events 261 76 241 71
Objective progression 206 60 197 58
Death without objective progression 55 16 44 13
Median, months† 4.7 4.6 1.10
95% CI 4.2 to 5.4 4.2 to 5.4 0.93 to 1.32
Best overall response (investigator assessment)¶ .68储
Complete response 2 0.6 3 0.9
Partial response 111 32 114 34
Stable disease 126 37 120 35
ORR 33 35
95% exact CI 28 to 38 29 to 40
Abbreviation: ORR, objective response rate. ⴱTwo-sided stratified log-rank test. †Brookmeyer and Crowley method.
‡Stratified Cox proportional hazards model v arm B. §Kaplan-Meier method.
¶Confirmed no sooner than 4 weeks after initial observation. 储Pearson2test.
ⱖ 5.7% than in those with levels less than 5.7% in both treatment arms. Median OS was approximately 1.5 months shorter in the figitumumab arm than in the control arm, regardless of baseline HbA1c(HR: patients with low HbA1c, 1.07; patients with high levels, 1.26).
Beyond the failure to demonstrate efficacy, a worrisome finding of this study was the relatively high frequency of treatment-related deaths associated with figitumumab (5%), an effect that was not detected in the phase II study. There are a number of potential reasons that may provide insight for future clinical trial design, dosing levels, and anticipation and management of toxicities, particularly where combination regimens are involved. First, only about half of the 98 patients randomly assigned to figitumumab in the phase II trial re-ceived the 20 mg/kg dose, a sample that might have been too small to
detect safety signals. However, the incidence of grade 3/4 hyperglyce-mia was greater in the phase II study (20%) than in our current study (12%).8Second, there were inherent differences in the patient popu-lations. For example, the phase III study enrolled patients with pre-dominantly squamous cell histology and far more current smokers (42% v 13%) than the phase II study. Hence, the phase III patient population may have had more attendant comorbidities (latent or overt), which might have rendered them more vulnerable to toxicity or intercurrent grade 5 events. Third, the phase II trial was conducted almost exclusively at tertiary referral centers, which may have led to subtle differences in the types of patients enrolled and how they were managed. As several study centers in the phase III trial enrolled only a few patients each, the investigators may have initially lacked
B
A
0 Overall Survival (%) Time (months) 100 80 60 40 20 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 l o r t n o C b a m u m u t i g i F Median OS, 8.6 (7.4 to 9.3) 9.8 (8.6 to 10.9) months (95% CI) HR (95% CI) 1.18 (0.99 to 1.40) P .06 Figitumumab Control 0 Progression-Free Survival (%) Time (months) 100 80 60 40 20 2 4 6 8 10 12 14 16 18 20 22 24 26 l o r t n o C b a m u m u t i g i F Median PFS, 4.7 (4.2 to 5.4) 4.6 (4.2 to 5.4) months (95% CI) HR (95% CI) 1.10 (0.93 to 1.32) P .27 Figitumumab Control No. at risk Figitumumab 342 279 221 174 131 100 81 71 48 35 15 4 1 0 Control 339 294 253 199 154 118 103 89 59 39 15 3 0 No. at risk Figitumumab 342 231 127 38 17 8 6 2 2 1 0 Control 339 236 114 47 21 12 8 4 0Fig 2. Kaplan-Meier estimates of (A) overall survival (OS) and (B) progression-free survival (PFS) in the randomly assigned population. HR, hazard ratio.
Favors figitumumab Favors control
Factor n HR 95% CI Natural log HR (95% CI)
Overall 681 1.18 0.99 to 1.40 Sex Female 159 1.09 0.75 to 1.60 Male 522 1.20 0.99 to 1.46 ECOG PS 0 228 0.99 0.73 to 1.34 1 443 1.21 0.98 to 1.50 Nonsquamous Yes 79 1.18 0.71 to 1.96 No 602 1.16 0.96 to 1.39 Smoker Never 67 1.46 0.83 to 2.57 Current 283 1.11 0.85 to 1.45 Stage IIIB 78 1.28 0.75 to 2.19 IV 602 1.16 0.97 to 1.40 HbA1c < 5.7% 267 1.07 0.81 to 1.41 ≥ 5.7% 371 1.26 1.00 to 1.60 –0.6 –0.4 –0.2 0 0.2 0.4 0.6 0.8 1.0 1.2
Fig 3. Forest plot of overall survival by
selected baseline characteristics. ECOG PS, Eastern Cooperative Oncology Group performance status; HbA1c, glycosylated hemoglobin; HR, hazard ratio.
Table 4. Most Common Treatment-Emergent (all-causality) Adverse Events for ⱖ 10% of Patients (any grade) or ⬎ 5% of Patients (grade 3 or 4) Adverse Event Grade 3 Grade 4 Any Grade Figitumumab Arm (n ⫽ 338) Control Arm (n ⫽ 333) Figitumumab Arm (n ⫽ 338) Control Arm (n ⫽ 333) Figitumumab Arm (n ⫽ 338) Control Arm (n ⫽ 333) No. of Patients % No. of Patients % No. of Patients % No. of Patients % No. of Patients % No. of Patients % Any adverse event (⬍ grade 5) 135 40 106 32 103 30 91 27 319 94 306 92 Alopecia 5 1 5 2 1 ⬍ 1 2 1 138 41 146 44 Nausea 12 4 2 1 0 0 133 39 103 31 Decreased appetite 23 7 7 2 0 0 130 38 75 23 Fatigue 19 6 11 3 2 1 2 1 112 33 86 26 Diarrhea 13 4 3 1 2 1 0 101 30 45 14 Anemia 15 4 20 6 4 1 0 95 28 88 26 Vomiting 9 3 3 1 0 0 85 25 47 14 Hyperglycemia 35 10 2 1 7 2 0 79 23 17 5 Neutropenia 20 6 31 9 45 13 33 10 77 23 78 23 Asthenia 23 7 15 5 4 1 1 ⬍ 1 7 52 2 6 11 8 Cough 5 1 4 1 0 0 66 20 60 18 Weight decreased 11 3 1 ⬍ 10 0 6 6 2 0 2 9 9 Peripheral neuropathy 13 4 14 4 0 2 1 62 18 57 17 Thrombocytopenia 17 5 15 5 10 3 5 2 61 18 52 16 Constipation 3 1 2 1 0 0 60 18 61 18 Dyspnea 8 2 15 5 4 1 3 1 60 18 68 20 Arthralgia 2 1 0 0 0 47 14 57 17 Peripheral sensory neuropathy 9 3 6 2 0 0 40 12 50 15 Dehydration 17 5 1 ⬍ 1 2 1 0 39 12 12 4 Myalgia 1 ⬍ 1 0 0 0 39 12 44 13 Dizziness 0 0 0 0 38 11 32 10 Hemoptysis 3 1 1 ⬍ 10 0 3 8 1 1 2 6 8 Headache 1 ⬍ 12 10 0 3 7 1 1 2 2 7 Pain in extremities 2 1 2 1 0 1 ⬍ 13 4 1 0 2 4 7 Paresthesia 1 ⬍ 11 ⬍ 1 0 0 2 993 5 1 1 Pyrexia 0 0 0 0 29 9 35 11 Leukopenia 8 2 14 4 4 1 4 1 19 6 34 10 Febrile neutropenia 3 1 12 4 3 1 6 2 7 2 18 5 NOTE. A total of 10 randomly assigned patients did not receive any study treatment and were not included in the safety analyses. One patient randomly as signed to the figitumumab arm received only paclitaxel and carboplatin and was included in the control arm in this analysis.
experience in managing figitumumab complications such as hyper-glycemia and dehydration.
Furthermore, although patient and disease characteristics were well balanced between arms, we cannot exclude the possibility that minor baseline demographic imbalances in this study might have produced inconsistent results. Although figitumumab was combined with the same full-dose chemotherapy doublet as in the phase II study, lower doses of the doublet might have improved the tolerability of the combination. Finally, the difference in the incidence of treatment-related grade 5 AEs between the figitumumab and control arms (5% v 1%), may indicate that signaling through IGF-1R and its attendant pathways is critical in maintaining homeostasis, such that inhibition significantly disrupts the insulin receptor/IGF-1R/growth-hormone signaling axis. This concern may be heightened in patients with ad-vanced squamous cell NSCLC, who often have multiple comorbidities and who constituted the vast majority of participants in this trial. Our experience highlights the potential discrepancies between phase II and phase III trials in both safety and efficacy, and underscores the impor-tance of identifying a priori the patient population(s) most likely to benefit from therapy. However, as seen in our current study, inclusion of such a selected patient group (predominantly squamous cell NSCLC) does not guarantee improved safety or efficacy.
In conclusion, though the phase II trial suggested an ORR advan-tage for adding figitumumab to standard chemotherapy in advanced NSCLC, our current phase III study involving nonadenocarcinoma patients failed to show any benefit and unexpectedly suggested a possible detrimental effect. This may be a class effect and should be assessed in current and future trials examining IGF-1R inhibitors. Further clinical development of figitumumab is not being pursued.
AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
Although all authors completed the disclosure declaration, the following author(s) and/or an author’s immediate family member(s) indicated a
financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a “U” are those for which no compensation was received; those relationships marked with a “C” were compensated. For a detailed description of the disclosure categories, or for more information about ASCO’s conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.
Employment or Leadership Position: Rebecca J. Benner, Pfizer (C);
Judith R. Scranton, Pfizer (C) Consultant or Advisory Role: Corey J. Langer, Pfizer (C); Keunchil Park, Boheringer Ingelheim (U), Eli Lilly (U), Roche (U); Daniel D. Karp, Pfizer (U); Tony Mok, AstraZeneca (C), Roche (C), Eli Lilly (C), Merck Serono (C), Eisai (C), Bristol-Myers Squibb (C), BeiGene (C), AVEO Pharmaceuticals (C), Pfizer (C), Taiho Pharmaceutical (C), Boehringer Ingleheim (C), GlaxoSmithKline (C)
Stock Ownership: Rebecca J. Benner, Pfizer; Judith R. Scranton, Pfizer Honoraria: Silvia Novello, Eli Lilly, Boehringer, Roche; Keunchil Park,
Eli Lilly, Roche, AstraZeneca; Tony Mok, AstraZeneca, Roche, Eli Lilly, Merck Serono, Eisai, Bristol-Myers Squibb, BeiGene, AVEO
Pharmaceuticals, Pfizer, Taiho Pharmaceutical, Boehringer Ingleheim, GlaxoSmithKline Research Funding: Corey J. Langer, Pfizer; Daniel D. Karp, Pfizer; Tony Mok, AstraZeneca Expert Testimony: None Patents,
Royalties, and Licenses: None Other Remuneration: None
AUTHOR CONTRIBUTIONS
Conception and design: Corey J. Langer, Tony Mok, Rebecca J. Benner,
Judith R. Scranton, Anthony J. Olszanski, Jacek Jassem
Provision of study materials or patients: Corey J. Langer, Silvia Novello,
Keunchil Park, Daniel D. Karp
Collection and assembly of data: Keunchil Park, Maciej Krzakowski,
Tony Mok, Rebecca J. Benner, Judith R. Scranton, Anthony J. Olszanski
Data analysis and interpretation: Corey J. Langer, Silvia Novello,
Keunchil Park, Daniel D. Karp, Tony Mok, Rebecca J. Benner, Judith R. Scranton, Anthony J. Olszanski, Jacek Jassem
Manuscript writing: All authors Final approval of manuscript: All authors
REFERENCES
1. National Cancer Institute: PDQ Non-Small Cell
Lung Cancer Treatment. Bethesda, MD, National Cancer Institute. http://www.cancer.gov/cancertopics/pdq/ treatment/non-small-cell-lung/healthprofessional (ac-cessed October 20, 2011)
2. Pollak M: Targeting insulin and insulin-like
growth factor signalling in oncology. Curr Opin Phar-macol 8:384-392, 2008
3. Fidler MJ, Shersher DD, Borgia JA, et al:
Targeting the insulin-like growth factor receptor pathway in lung cancer: Problems and pitfalls. Ther Adv Med Oncol 4:51-60, 2012
4. Favoni RE, de Cupis A, Ravera F, et al:
Ex-pression and function of the insulin-like growth factor I system in human non-small-cell lung cancer and normal lung cell lines. Int J Cancer 56:858-866, 1994
5. Kaiser U, Schardt C, Brandscheidt D, et al:
Expression of insulin-like growth factor receptors I and II in normal human lung and in lung cancer. J Cancer Res Clin Oncol 119:665-668, 1993
6. Haluska P, Shaw HM, Batzel GN, et al: Phase
I dose escalation study of the anti insulin-like growth factor-I receptor monoclonal antibody CP-751,871 in patients with refractory solid tumors. Clin Cancer Res 13:5834-5840, 2007
7. Lacy MQ, Alsina M, Fonseca R, et al: Phase
I, pharmacokinetic and pharmacodynamic study of the anti-insulinlike growth factor type 1 receptor monoclonal antibody CP-751,871 in patients with multiple myeloma. J Clin Oncol 26:3196-3203, 2008
8. Karp DD, Paz-Ares LG, Novello S, et al:
Re-traction: Phase II study of the anti–insulin-like growth factor type 1 receptor antibody CP-751,871 in combination with paclitaxel and carboplatin in previously untreated, locally advanced, or metastatic non-small-cell lung cancer. J Clin Oncol 30:4179, 2012
9. American Joint Committee on Cancer: AJCC
Cancer Staging Manual (ed 6). New York, NY, Springer, 2002, pp 167-181
10. Therasse P, Arbuck SG, Eisenhauer EA, et
al: New guidelines to evaluate the response to treatment in solid tumors: European Organisation
for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 92:205-216, 2000
11. Sobin LH, Wittekind C (eds): International
Union Against Cancer (UICC), TNM Classification of Malignant Tumours (ed 6). New York, NY, Wiley-Liss, 2002
12. Goto Y, Sekine I, Tanioka M, et al:
Figitu-mumab combined with carboplatin and paclitaxel in treatment-naïve Japanese patients with advanced non-small cell lung cancer. Invest New Drugs 30: 1548-1556, 2012
13. Murakami H, Doi T, Yamamoto N, et al: Phase
1 study of ganitumab (AMG 479), a fully human monoclonal antibody against the insulin-like growth factor receptor type I (IGF1R), in Japanese patients with advanced solid tumors. Cancer Chemother Pharmacol 70:407-414, 2012
14. Gualberto A, Pollak M: Emerging role of
insulin-like growth factor receptor inhibitors in oncol-ogy: Early clinical trial results and future directions. Oncogene 28:3009-3021, 2009
Acknowledgment
We thank the participating patients and their families, as well as the network of investigators (Appendix), research nurses, study coordinators, and operations staff. We also posthumously acknowledge the contribution of Valentina Tzekova, MD, University Hospital Queen Joanna, Sofia, Bulgaria, to the study. We also thank Janos Strausz and Igor Bondarenko for their contributions to the study. Medical writing support
was provided by Nicola Crofts at ACUMED (Tytherington, United Kingdom) and was funded by Pfizer.
Appendix
Investigators. Charles A. Butts, MD, Daniele Marceau, MD, Christopher Charles Croot, MD, Shirish Madhav Gadgeel, MD, Mansoor Noorali Saleh, Lee C. Drinkard, MD, Haralambos E. Raftopoulos, MD, Isidoro Barneto Aranda, Luis Alfonso Gurpide Ayarra, Pilar Lopez Criado, Felipe Cardenal Alemany, Marta Lopez-Brea Piqueras, Amelia Insa Molla, Roy Timothy Webb, MD, Maciej Krzakowski, PhD, Kazimierz Roszkowski-Sliz, PhD, Janusz Rolski, MD, Tadeusz Tobiasz, MD, Petr Zatloukal, MD, Jitka Jakesova, MD, Martin Smakal, MD, Saime Ayse Kars, MD, PhD, Ismail Oguz Kara, MD, PhD, John Allan Ellerton, MD, Kurt Aigner, MD, Sabine Zoechbauer-Mueller, MD, PhD, Nashat Yousif Gabrail, MD, Hari Menon, MD, Anand Pathak, MD, Digambar Behera, MD, Charles Arthur Henderson, MD, Anthony Mark Landis, MD, Carlos Gil Moreira Ferreira, MD, Ulf Petrausch, MD, Daniel C. Betticher, PhD, Barna Szima, MD, Zsuzsanna Mark, MD, Zsolt Szekely Papai, MD, Attila Somfai, MD, Janos Strausz, MD, PhD, Adam Richard Broad, MD, Bogdan N. Kotiv, MD, Dmitriy P. Udovitsa, MD, Henrik Riska, MD, Matti Pietilainen, MD, Nathan Adam Pennell, William Graydon Harker, MD, Daniel David Karp, MD, Donald St Paul Gravenor, MD, Giorgio Vittorio Scagliotti, PhD, Francesco Grossi, MD, Davide Pastorelli, MD, Filippo De Marinis, PhD, Alain Catalin Mita, MD, Everett Emmett Vokes, MD, Jose Rodrigues Pereira, MD, Violina Taskova, MD, Constanta Velinova Timcheva, PhD, Krassimir Dimitrov Koynov, MD, Tatyana Vasileva Koynova, MD, Wieslaw Wiktor Jedrzejczak, PhD, Dae-Seog Heo, MD, Keunchil Park, MD, Heung Tae Kim, MD, Peter Berzinec, Pavol Demo, MD, Ewa Jankowska, MD, Michael T. Slaughter, MD, Tony S.K. Mok, PhD, Tsai Chun-Ming, Su Wu-Chou, MD, Meng-Chih Lin, MD, Chih-Hsin Yang, MD, Joseph Thomas Meschi, MD, Konstantinos Syrigos, PhD, George Fountzilas, PhD, Michael Vaslamatzis, MD, Stephen Lloyd Graziano, MD, Ping Fai So, MD, Frank Griesinger, MD, PhD, Martin Reck, MD, Martin Sebastian, MD, Sylvia Guetz, MD, Dennis Eli Slater, MD, Jean-Claude Guerin, Jean Yves Douillard, PhD, Claude El Kouri, MD, Herve Lena, MD, Patrick Merle, MD, Gerard Zalcman, PhD, Joerg Mezger, MD, PhD, Fabrice Paganin, MD, Christos Emmanouilides, MD, Sandip Abhay Shah, MD, Ganesha D. Vashishta, MD, Yaroslav Shparyk, MD, Nataliya L. Voytko, MD, Igor Mykolayovych Bondarenko, PhD, Vera Andreevna Gorbunova, PhD, Vasily I. Borisov, PhD, Oleksandr Yu. Popovych, PhD, Igor O. Vynnichenko, MD, Goetz H. Kloecker, MD, James Patrick Daugherty, MD, Janak K. Choksi, MD, Troy Hancil Guthrie Jr, MD, Sing Hung Lo, MD, Stephen Begbie, MD, Haluk Tezcan, MD, Susan Amy Sajer, MD, David William Zenk, MD, Samuel Spence McCachren Jr, MD, Stephen Daniel Myers, MD, Mark Ramsey Hutchins, MD, Richard Scott Siegel, Eric Powell Lester, MD, Tarek Eldawy Mohamed, MD, Erwin Lee Robin, MD, Nicholas Oswald Iannotti, MD, Ahmad Ali Tarhini, MD, Corey Jay Langer, MD, Steven Charles Buck, MD, CheolWon Suh, MD, Brian Nicholson Mathews, MD, David Holden Henry, MD, Manuel Francisco Gonzalez, MD, Juraj Beniak, MD, Robert Matthew Jotte, MD, Michaela Long Tsai, MD, William Connelly Waterfield, MD, David A. Van Echo, MD, Waseemullah Khan, MD, Gary E. Goodman, MD, John Ward McClean, MD, Ronald George Steis, MD, Konstantin Hristov Dragnev, MD, Lee M. Zehngebot, MD, Hector A. Velez Cortes, Stacey Kay Knox, MD, Sergey V. Orlov, MD, Mikhail V. Kopp, MD, Bruno Coudert, MD, Radj Gervais, MD, Robert Harold Gersh, MD, Michael Alan Savin, MD, Rama Koteswararoa Koya, MD, Marcus Alan Neubauer, MD, Alexander Illya Spira, MD, Jacek Jassem, PhD, Edward Thomas O’Brien, MD, Linda L. Ferris, MD, Richard C. Frank, MD, Beth Ann Hellerstedt, MD, Donald Anthony Richards, MD, Takeshi Horai, MD, Noboru Yamamoto, MD, Hiroshi Isobe, MD, Miyako Satouchi, MD, Shinji Atagi, MD, Isamu Okamoto, MD, Toshiyuki Sawa, MD, Hiroaki Okamoto, MD, Koji Takeda, MD, Tetsu Shinkai, MD, Richard Wilhelm Eek, MD, Valentina Ilieva Tzekova, PhD, Susan Fox, MD, Jan Novotny, MD, Maurice Perol, MD, Masaaki Kawahara, MD, Pilar Lopez Criado, Raul Manuel Marquez Vazquez, Susanna Stoll, MD, Alessandra Curioni, MD, Ismail Oguz Kara, PhD, Sinan Yavuz, MD, PhD, Tarek Mouris Mekhail, MD, Athanassios Argiris, MD, Brian Vincent Geister, MD, Samir Ezzeldin Witta, MD, Joseph Ward Leach, MD, and Nicholas James Robert, MD.
Table A1. Boundaries for the Planned Interim Analyses and First Interim Analysis Results
Analysis
No. of Events
Efficacy Boundaries for Rejecting the Null Hypothesis
Futility Boundaries for Rejecting the Alternative Hypothesis Z Score Hazard Ratioⴱ Z Score Hazard Ratioⴱ
Plan for first interim analysis 216 3.710 0.603 ⫺0.695 1.10
Plan for second interim analysis 433 2.511 0.785 1.003 0.908
Plan for final analysis 649 1.993 0.855
Computed boundaries for actual first interim analysis 225 3.632 0.616 ⫺0.595 1.083
Observed results for first interim analysis 225 ⫺1.407† 1.209†
ⴱHazard ratio was computed from z score boundary assuming proportional hazards and approximating the SE using the No. of events. Decisions were based on the z score boundaries.
Table A2. Grade 5 All-Causality Adverse Events, Excluding Disease Progression
Adverse Event
Figitumumab Arm (n⫽ 338) Control Arm (n⫽ 333)
No. of Patients % No. of Patients %
Totalⴱ 43 13 32 10 Pulmonary hemorrhage 7 2 4 1 Pneumonia 7 2 3 1 Cardiopulmonary failure 5 1 2 1 Sepsis 4 1 1 ⬍ 1 Death 3 1 3 1 Pulmonary embolism 1 ⬍ 1 3 1 Respiratory failure 2 1 1 ⬍ 1 Renal failure 2 1 1 ⬍ 1 Cerebrovascular accident 0 3 1 Gastrointestinal hemorrhage 0 2 1 Cardiovascular disorder 2 1 0 Hemorrhage 2 1 0 Sudden death 1 ⬍ 1 1 ⬍ 1
Performance status decreased 1 ⬍ 1 1 ⬍ 1
Arrhythmia 1 ⬍ 1 0 Emphysema 1 ⬍ 1 0 Hypovolemic shock 1 ⬍ 1 0 Hypoxia 1 ⬍ 1 0 Myocardial infarction 1 ⬍ 1 0 Multiorgan failure 1 ⬍ 1 0 Neutropenia 1 ⬍ 1 0 Neutropenic sepsis 1 ⬍ 1 0 Staphylococcal infection 1 ⬍ 1 0
Toxicity to various agents 1 ⬍ 1 0
Asphyxia 0 1 ⬍ 1 Cardiac failure 0 1 ⬍ 1 Dehydration 0 1 ⬍ 1 Dyspnea 0 1 ⬍ 1 Pulmonary edema 0 1 ⬍ 1 Suicide 0 1 ⬍ 1
Superior vena cava syndrome 0 1 ⬍ 1
ⴱP⫽ .22 by two-sided Fisher’s exact test. Some patients had more than one grade 5 adverse event.
B
A
0 Overall Survival (%) Time (months) 100 80 60 40 20 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 l o r t n o C b a m u m u t i g i F ) 2 8 1 = n ( ) 2 9 1 = n ( Median OS, 7.0 (5.8 to 8.2) 10.1 (8.0 to 11.4) months (95% CI) HR (95% CI) 1.37 (1.09 to 1.73) P .01 Figitumumab Control Overall Survival (%) Time (months) 100 80 60 40 20 l o r t n o C b a m u m u t i g i F ) 6 9 = n ( ) 0 2 1 = n ( Median OS, 10.4 (9.0 to 12.0) 9.4 (8.6 to 11.1) months (95% CI) HR (95% CI) 0.93 (0.67 to 1.29) P .67 Figitumumab Control No. at risk Figitumumab 193 153 116 86 65 52 41 36 19 13 6 0 Control 182 157 138 108 88 67 60 51 30 21 6 2 0 No. at risk Figitumumab 120 100 82 73 56 42 36 31 26 19 8 4 1 0 Control 96 85 72 57 40 33 27 23 16 10 5 1 0Fig A1. Kaplan-Meier estimates of overall survival (OS) in patients with (A) total baseline insulin-like growth factor 1 (IGF-1) less than 120 ng/mL or (B) total baseline
